The rapid and accurate identification of bacteria has been considerably advanced by the development of commercial, mainly manual systems. Automated systems are being developed which will further facilitate these identification procedures. A specially designed, commercially prepared microdilution plate is being read by a commercial reading device and a large data base has been developed. Mathematical modeling of the data base has demonstrated a diagnostic accuracy of approximately 96%. Initial steps have been completed to interface this automated reading instrument with the laboratory computer. The development of this interface, and the routine use of the automated reader, will allow the laboratory to streamline its identification procedures and to greatly reduce transcription steps to get data into the laboratory computer. A new, more rapid model of the above instrument is now available and is being exchanged for the older model. This will require some additional software modifications, but the increased computer power and rapidity of operation will make these changes cost effective.